In the days of the mainframe, systems programmers kept odd hours. I guess we still do. People are apt to get emails from me at all hours. Today is a lot easier to work on systems code, though, because back then we only had a couple of machines. Getting time on one of those machines was limited to the wee hours of the morning.
After all the day’s jobs had run, after the evening’s business, admissions, and library updates had run, even after the RACF clean had run, long after all the tapes stopped being mounted, and the printers fell silent, only then could the systems programmers emerge from our offices to make our systems changes.
We mounted our own tapes, since all the operators but one had gone home. We would submit our batch jobs from TSO and go to the tape alley to watch the JES console for our job name to pop up. MOUNT TAPE UPDATE21 ON UNIT 7, it would read. We would take our tape to the head high tape transport labeled with a large 7.
To get to the tape spindle, we’d snap back the top edge of the clear glass panel that protected the take-up reel. After removing the write-rings, we put our tapes, all of which had autoloading rings, on the spindle, snapped the latch closed, and hit the load button. With a zip and a loud rush of air, the glass would rise to the top, and the end of the tape would swoosh through a small hole.
It would blow across and get caught up in the take-up reel as if by some magic hand. Air is an amazing thing when used right. With a strange honking sound, the tape would be sucked down into the two four-foot vacuum columns (two loops) used to keep the supply and take reels positioned so the capstan could quickly accelerate and decelerate.
A tape could go from a dead stop to full speed in six-tenths of an inch. That’s what vacuum columns can do for you.
As soon as the end of the tape was around the take-up reel, it would speed forward until it found a one inch by quarter inch piece of shiny silver mylar - the beginning of tape marker. It would then be ready to go.
This was often a tape crammed with a huge amount of code, maybe even all of 40 MB (yes, that’s a M as in million and the B is bytes). It may have been an update to some part of the MVS system, or maybe part of a compiler. We used software from NASA that was cleared by a group called Cosmic. Patches to be applied, software systems to be rebooted, a quick check to verify all was good again, and the elves would depart for the day.
Like the mists of the morning, us system programmers vanished. Others would find our cubicles empty and ask, “Where are they?” “Updated the CISC system last night.” “Oh! Better them than me.”
Such was the life of the budding systems programmer in the late 1970s.
Then one night, the drowsy operator in his white wrinkly shirt, stirred from his usual slouched position, half asleep watching the late night jobs go through the system.
His eyes opened, his head popped up slightly from his palm where it has been resting. Then he shifted his chair, squinting at the screen.
“What the heck?” he mumbled. “Hey guys, come look at this.”
Slowly we stirred from our own drowsy duties and made our way to the group of 3270 terminals, which were our operator consoles. The printers were in front of us, the disk and tape alleys behind. Beyond the printers was the wall.
You could say the wall had 36 pigeon holes (if the pigeons were the size of turkeys). In these boxes, the fruit of the five large 1403 printers would be separated and placed in the hole with the first letter or number of the job name. I knew the “B” box very well.
“What are you guys doing back there?” the operator asked of us.
When we got to the operator terminal, we could see that it had cleared. Instead of the usual series of green letters in a sea of black (letters in a font only IBM could design—I mean who puts tiny little dots in their zeros?), there were two numbers: 2:47. As the second number ticked down, we realized it was a countdown.
“Man, that’s nothing we’re doing back there,” I said, pushing my long hair out of my eyes. The operator, a heavy man with a crew cut, did not believe me. He insisted that it was something we were doing.
As it rolled to 1:59, we knew it was a timer.
For a few moments every science fiction, spy thriller, and television show with a bomb counting down was playing in our heads. Maybe it was the late hour, maybe the fatigue, but for some seconds the room spun a little as I wondered how someone could get two four-million dollar machines to explode!
My buddies ran to another terminal to try to see what job was running. What job had taken over the VTAM controller to preempt the usual mundane lines of JES2 messages in order to display a countdown timer?
We held our breath as it counted down. It went to 0:59. It continued to countdown. As it did, we stopped breathing, frozen in disbelief, part scared, part dismissive, part whistling in the dark.
Then it went black. After a few seconds, it said “STEP IN THE BOX!”
Step in the box?
As we thought about what that must mean, one of the compact-car-sized printers started to spin up. They worked with a massive print chain that looped around the front of the paper. When you sent something to the printer, the chain would spin up to speed before a row of 132 hammers behind the paper and carbon would strike at just the right letter in just the right place at just the right time.
These beasts could print this way at a rate of 600 lines a minute. It was impressive. You heard the entire 132 character line go WHAM, WHAM, WHAM as each line was banged out on the printer.
And they weren’t just fast. Each one could slew the paper to the top of the next page at a terrifying rate. These beasts were so loud, they had to be encased in a cabinet of soundproofing with a little window to collect printouts through. The cabinet was so heavy, it had a built-in power lift so we could raise the lid to the green paper it used, the kind that was fed by little pinwheels.
One of those beasts was spinning up. Then, all of a sudden, we heard a loud WHOOSH.
“Damn it!” the operator grunted as he jumped up and ran to the printer. “The damn thing is slewing all the paper out!’’
And it was. By the time he got there, only a few feet away, the paper box was empty, all the blank paper piled up in a mess where it would normally be sitting in wait to be split apart and sorted by job names onto the little wooden shelves.
That was my first virus. It was during the Carter administration, though I don’t think that had anything to do with it.
The virus struck a few more times. The timer started with smaller values later, giving us only 60 seconds to notice and get to the printer. It would randomly pick one, so we didn’t know until the printer would start to spin up which printer was about to empty a box of paper onto the floor.
The fix, or workaround, I should say, was to hit the open button on the hood and step in the box of paper, breaking it off. We had to STEP IN THE BOX!
It didn’t happen every night, just once in a while. For about a week after the first one, the night operator would be very alert. Then, after a while, they would get complacent, they would get sleepy, they would again assume the position with their head in hand, their elbow on the table, and whamo. It would happen again.
Until one day we realized it hadn’t happened in a while. It had stopped. We don’t know why. Maybe it had a counter and would move on, written by some mischievous system programmer with an ounce of compassion. Maybe the part of the operating system it was embedded in had been updated and accidentally removed from the system. In any event, it had disappeared.
But, still, to this day, the words STEP IN THE BOX, get my attention.